@thebassclef/lite 0.1.2 → 1.0.0

package/substrate/standards/project-directory-layout.md

@@ -1,276 +0,0 @@
----
-tier: lite
-description: Standard directory structure for sunj-labs repositories.
----
-
-# Project Directory Layout
-
-Standard directory structure for sunj-labs repositories. The goal: a new adopter cloning a bassclef-consumer repo can locate any artifact in under 30 seconds without asking.
-
-> **Which page do I want?** Two related pages cover "where does X go?":
->
-> - **This page (Directory Layout)** — the methodology tree. Answers: "I'm writing a canvas / spec / ADR / chronicle — which directory does it live in?" Use when you're producing an SDLC artifact and need to know where to put it.
-> - **[File structure](https://github.com/sunj-labs/bassclef/wiki/File-structure)** — the installation footprint. Answers: "What did `/onboard-repo` actually write into my repo? What's safe to edit vs. bassclef-managed?" Use when you're inspecting what landed after install or deciding what to commit vs. gitignore.
->
-> If you're asking "where does this artifact go?" you want Directory Layout. If you're asking "what just got installed?" you want File structure.
-
-This standard replaces the prior 2026-01 snapshot. See `docs/audits/2026-05-12-directory-skill-overlap-audit.md` for the inventory that motivated this rewrite.
-
-## Mental model
-
-Every artifact has one home. The SDLC produces artifacts in a predictable flow:
-
-```
-idea → canvas → spec → design → prototype → build → chronicle
-       (strategy/) (specs/) (designs/) (prototypes/) (src/) (chronicle/)
-```
-
-### Two repo shapes — why your app repo and bassclef look almost identical except for `docs/`
-
-The two layouts you'll see on this page (App repo and Bassclef) have the same SDLC artifact categories. The only structural difference is where those categories live:
-
-- **App repo** (yours): SDLC artifacts nest under `docs/` so they don't compete with `src/` at the top. Your application code is what readers expect to see first when they open the repo.
-- **Bassclef** (the methodology repo): the entire repo IS the methodology. There is no `src/` at the top. SDLC artifact categories hoist to root because that's the load-bearing content.
-
-Same artifacts, different nesting. A canvas in your app repo lives at `docs/strategy/canvases/`. A canvas in bassclef lives at `strategy/canvases/`. The path-prefix is the only delta.
-
-| Artifact | App repo (your repo) | Bassclef (this repo) |
-|---|---|---|
-| Canvas | `docs/strategy/canvases/` | `strategy/canvases/` |
-| Spec | `docs/specs/` | `docs/specs/` (bassclef happens to use docs/specs too) |
-| ADR | `docs/architecture/decisions/` | `architecture/decisions/` |
-| C4 diagram | `docs/architecture/c4/` | `architecture/c4/` |
-| Design tokens | `docs/designs/` | `design/` |
-| Chronicle | `docs/chronicle/` | `chronicle/` |
-| Decomposition | `docs/decompositions/` | `docs/decompositions/` |
-| Iteration goal | `docs/iteration-bets/` | `docs/iteration-bets/` |
-
-The asymmetry is intentional, not legacy. App repos need `src/` at the top; bassclef doesn't have one to compete with. Same conceptual flow (`idea → canvas → spec → ...`), different root-level placement.
-
-## App repo layout
-
-Legend:
-
-- 🔵 **bassclef-managed** — the substrate writes/owns these paths. Skills, rules, hooks, agents, and SDLC artifacts produced by the methodology.
-- ⚪ **your code** — your application. Bassclef doesn't touch `src/`, your migrations, your tests. Names below are illustrative; replace with your stack's conventions.
-
-```
-⚪ src/                            # your application code
-⚪ your-database-migrations/       # e.g. prisma/, alembic/, knex/ — if applicable
-⚪ public/                         # your static assets
-⚪ scripts/                        # your build / deploy / utility scripts
-⚪ tests/                          # your test suites (unit / integration / e2e)
-⚪ data/                           # your data files, snapshots (not docs)
-
-🔵 .claude/                        # bassclef substrate (skills, rules, hooks, agents)
-
-🔵 docs/                           # SDLC artifacts produced by the methodology
-🔵   strategy/
-🔵     canvases/                   # product canvases (Thesis → Shape → Build)
-🔵     pr-faqs/                    # PR/FAQ documents
-🔵     research/                   # research notes
-🔵   specs/                        # structured specs from canvases
-🔵   decompositions/               # /decompose output (responsibility audit + pattern map)
-🔵   designs/                      # design artifacts (tokens, mocks, references)
-🔵   ia-models/                    # information architecture models
-🔵   interaction-design/           # interaction design artifacts
-🔵   prototypes/                   # visual + technical prototypes (HTML, screenshots, gallery indexes)
-🔵   visual-reviews/               # /visual-review output
-🔵   architecture/
-🔵     decisions/                  # ADRs
-🔵     c4/                         # C4 diagrams
-🔵     erd/                        # entity-relationship diagrams
-🔵     reviews/                    # /architect-review output
-
-🔵   # iteration + roadmap
-🔵   roadmaps/                     # multi-week strategic frames (compose N bets)
-🔵   iteration-bets/               # time budget + scope per iteration
-🔵   branch-stacks/                # stack manifests for multi-branch work
-
-🔵   # session lifecycle artifacts
-🔵   chronicle/                    # session narratives (dated, append-only)
-🔵   sessions/                     # session-board coordinator artifacts (multi-bet longruns)
-🔵   sdlc-traces/                  # session hook logs
-🔵   deferred-actions/             # cross-session handoff entries
-🔵   audits/                       # audit reports (adopter-readiness, directory drift, etc.)
-🔵   sibling-validations/          # validation reports from sibling-repo runs
-
-🔵   # comms + ops
-🔵   journals/              # journal-entry drafts (operator narrative voice; publishable to LinkedIn, Substack, etc.)
-🔵   journal-export/              # journal-entry corpus exports to brand-team Google Doc
-🔵   release-notes/                # per-release notes
-🔵   demos/                        # demo runbooks
-🔵   runbooks/                     # operational runbooks
-🔵   research/                     # operator research notes
-🔵   learnings/                    # captured learnings (sister to design/discoveries in bassclef)
-🔵   longrun-plans/                # /longrun pre-flight artifacts
-🔵   metrics/                      # operator-facing metrics dashboards
-```
-
-Stack-specific entries (database migrations, framework public folders, build tool config) live alongside `src/` per your stack's conventions — bassclef doesn't move your code. The 🔵 rows are what the substrate writes; the ⚪ rows are placeholders showing where your code lives relative to the substrate.
-
-Optional `docs/strategy/` subdivision matches bassclef's `strategy/` root — same artifact types, different nesting.
-
-## Bassclef (methodology repo) layout
-
-```
-.claude/                      # bassclef's substrate (skills, rules, hooks, agents)
-architecture/                 # ADRs + C4 + ERD + reviews (hoisted from docs/architecture/)
-  decisions/
-chronicle/                    # session narratives (hoisted from docs/chronicle/)
-design/                       # design principles, object model, tokens (hoisted from docs/designs/)
-  discoveries/                # earned-wisdom principles (bassclef-only)
-lib/                          # state-spine accessor library (lib/state.sh)
-patterns/                     # pattern catalog (GoF / Fowler / Smarter Patterns / etc)
-sdlc-portable/                # exportable SDLC bundles
-standards/                    # engineering standards (this file lives here)
-  state-spine/                # state-spine schemas
-state/                        # state-spine markers + projections (runtime)
-scripts/                      # bassclef-specific utility scripts
-strategy/                     # canvases + templates (hoisted from docs/strategy/)
-  canvases/
-  templates/
-tests/                        # bassclef hook tests
-docs/                         # everything else from the app-repo layout above
-  audits/
-  branch-stacks/
-  baselayer/                  # baselayer-publish strategy + queue (bassclef-specific)
-  decompositions/
-  deferred-actions/
-  demos/
-  designs/
-  iteration-bets/
-  ia-models/
-  input-artifacts/            # rendered input artifacts from /interpret-input
-  interaction-design/
-  interviews/                 # interview transcripts feeding canvases
-  learnings/
-  journals/
-  journal-export/
-  longrun-plans/
-  metrics/
-  release-notes/
-  research/
-  roadmaps/
-  runbooks/
-  sdlc-traces/
-  sessions/
-  sibling-validations/
-specs/                        # at bassclef root by exception (currently 1 file); migrate to docs/specs/
-```
-
-## One artifact, one location — the bassclef map
-
-Reference for adopters: "I have artifact X — where does it go?"
-
-| Artifact | App repo path | Bassclef path | Produced by |
-|---|---|---|---|
-| Canvas | `docs/strategy/canvases/` | `strategy/canvases/` | `/canvas` |
-| Spec | `docs/specs/` | (use `docs/specs/`; legacy `specs/` slated for removal) | `/spec` |
-| ADR | `docs/architecture/decisions/` | `architecture/decisions/` | `/architect-review`, manual |
-| C4 diagram | `docs/architecture/c4/` | `architecture/c4/` | `/architect-review` |
-| Decomposition | `docs/decompositions/` | `docs/decompositions/` | `/decompose` |
-| Design tokens / mocks | `docs/designs/` | `design/` | `/frontend-design`, manual |
-| IA model | `docs/ia-models/` | `docs/ia-models/` | `/ia-model` |
-| Interaction design | `docs/interaction-design/` | `docs/interaction-design/` | `/interaction-design` |
-| Prototypes | `docs/prototypes/` | `docs/prototypes/` | `/preview`, `/preview-build`, `/tech-prototype` (see "Skill family for prototypes") |
-| Visual review | `docs/visual-reviews/` | `docs/visual-reviews/` | `/visual-review` |
-| Chronicle | `docs/chronicle/` | `chronicle/` | `/chronicle`, `/session-end` |
-| Journal entry | `docs/operator-private/journals/` | `docs/operator-private/journals/` | `/journal` |
-| Release note | `docs/release-notes/` | `docs/release-notes/` | `/release-notes` |
-| Iteration goal | `docs/iteration-bets/` | `docs/iteration-bets/` | manual / `/longrun prep` |
-| Roadmap | `docs/roadmaps/` | `docs/roadmaps/` | manual |
-| Stack manifest | `docs/branch-stacks/` (SENTINEL_bassclef_name) | `docs/branch-stacks/` | `/longrun prep` (see "Stack manifest path migration") |
-| Session board | `docs/sessions/` | `docs/sessions/` | `/longrun prep` (multi-bet only) |
-| SDLC trace | `docs/sdlc-traces/` | `docs/sdlc-traces/` | hooks |
-| Deferred action | `docs/deferred-actions/` | `docs/deferred-actions/` | session-rescue, skills that defer |
-| Audit report | `docs/audits/` | `docs/audits/` | manual / agent audit runs |
-| Sibling validation | `docs/sibling-validations/` | `docs/sibling-validations/` | sibling-validation runs |
-| Demo runbook | `docs/demos/` | `docs/demos/` | manual |
-| Operational runbook | `docs/runbooks/` | `docs/runbooks/` | manual |
-| Research note | `docs/research/` | `docs/research/` | `/agent-research-spawn`, manual |
-| Learning | `docs/learnings/` | `docs/learnings/` | manual (bassclef: design/discoveries/ for earned wisdom) |
-| Metrics dashboard | `docs/metrics/` | `docs/metrics/` | `/metrics` |
-| Longrun plan | `docs/longrun-plans/` | `docs/longrun-plans/` | `/longrun prep` |
-| Input artifact | `docs/input-artifacts/` | `docs/input-artifacts/` | `/interpret-input` |
-| Interview transcript | `docs/interviews/` | `docs/interviews/` | manual / `/interview` (when shipped) |
-| Journal-entry corpus | `docs/journal-corpus/` | `docs/journal-corpus/` | `/journal-export` |
-| PR/FAQ | `docs/strategy/pr-faqs/` | `templates/pr-faq.md` (template only) | `/canvas` |
-
-## Skill family for prototypes — three bassclef surfaces (per ADR-020)
-
-Three skills, two axes. Adopter decision tree:
-
-| If you want | Use | Sizes |
-|---|---|---|
-| Visual mock variants only (clickable, on a subdomain) | `/preview` | quick (~30 min, 2 variants) / light (~2h, 3 variants + Anthropic baseline) |
-| Visual mock variants + buildable plan (spec + decomposition + migration) | `/preview-build` | medium (~1 day) / full (audit-grade) |
-| Pre-build risk gate (accuracy / integration / UX-visual / structural) | `/tech-prototype` | per shape |
-
-Two skills are deprecated per ADR-020 (2026-05-12):
-
-- `/prototype-variants` → use `/preview light` (light size already includes luminary champion-challenger)
-- `/prototype-gallery` → use `/preview` (any size; `/preview` already produces a shareable gallery URL per ADR-016)
-
-The deprecated skills remain invocable for ≥3 months while existing canvases / bets / chronicles that reference them get migrated. Full retirement schedule is in ADR-020 §"Migration plan."
-
-## Stack manifest path migration
-
-The directory layout standard names `docs/branch-stacks/` as bassclef. Reality has stack manifests split between `docs/branch-stacks/` (13 files) and `docs/stacks/` (36 files). The 36:13 split tilts toward the alias.
-
-**Decision (this version):** `docs/branch-stacks/` is the SENTINEL_bassclef_name; `docs/stacks/` is being migrated. New manifests land in `docs/branch-stacks/`. A follow-on migration relocates the 36 files in `docs/stacks/` and updates references.
-
-**Migration ticket:** TBD (filed after this standard merges).
-
-## Rules
-
-### One location per artifact
-
-Every artifact type has exactly one directory. No duplicates, no fallback chains. If you're unsure where something goes, check the bassclef map above.
-
-### Create on use, not upfront
-
-Don't scaffold empty directories. Create a directory when the first file needs it. The standard defines WHERE things go, not that every directory must exist from day one.
-
-### Add a new directory? Update this standard
-
-When a skill or workflow needs a new top-level directory under `docs/` (or bassclef's root), update the bassclef map in the same PR. Standard-vs-reality drift is the bug class this rewrite addresses; we don't want to grow another generation of it.
-
-### Naming conventions
-
-- Chronicles: `YYYY-MM-DD-slug.md`
-- Journal entries: `YYYY-MM-DD.md`
-- Iteration goals: `YYYY-MM-DD-slug.md`
-- Roadmaps: `YYYY-MM-DD-slug.md`
-- Canvases: `canvas-{name}.md` or `product-canvas.md`
-- ADRs: `ADR-NNN-short-description.md`
-- Specs: `SPEC-NNN-short-description.md` or `{feature-name}.md`
-- Session boards: `YYYY-MM-DD-slug.md`
-- Stack manifests: `YYYY-MM-DD-slug.md`
-- Audits: `YYYY-MM-DD-slug.md`
-- Sibling validations: `YYYY-MM-DD-slug.md`
-- Decompositions: `slug.md` (no date prefix; one per system/component)
-
-### Composition between artifacts
-
-Roadmaps, iteration goals, and stack manifests compose into a work-decomposition tree. Cross-references are typed in frontmatter, not implicit-by-prose. See `standards/artifact-composition.md`.
-
-### Brownfield retrofit
-
-When adopting bassclef in an existing repo:
-
-1. **Don't move existing application code.** Bassclef's substrate goes in `.claude/`; your app stays where it is.
-2. **Create `docs/` if it doesn't exist.** Bassclef wants its artifacts there.
-3. **Map existing methodology docs to the bassclef paths.** If you have `notes/` with iteration plans, move them to `docs/iteration-bets/`. If you have `architecture/` at root, move to `docs/architecture/` (app repo) or hoist to root (if bassclef-style methodology repo).
-4. **Add a one-time `docs/legacy/` subdirectory if needed.** Preserves pre-bassclef content with attribution; delete after migration completes.
-5. **Update your README.md** to point at `docs/` for the methodology layer.
-
-### What does NOT go in `docs/`
-
-- Application code (`src/`)
-- Database snapshots and data files (`data/`)
-- Build artifacts (`.next/`, `dist/`, `build/`)
-- Secrets and credentials (`.env` — never committed)
-- Node modules (`node_modules/`)
-- Playwright MCP artifacts (`.playwright-mcp/`, gitignored)
-- Substrate scripts (`.claude/hooks/`, `.claude/skills/`, etc — that's bassclef)

package/substrate/standards/release-tagging.md

@@ -1,137 +0,0 @@
----
-tier: lite
-description: Every release of bassclef carries a semantic version tag.
----
-
-# Release tagging
-
-Every release of bassclef carries a semantic version tag. The tag names what adopters can pin, roll back to, and read a changelog against. This standard names when tags cut, how versions bump, and what the release notes carry.
-
-Bassclef ships to public bassclef via `/release` PRs (per ADR-029). Before this standard, the substrate carried no version tags — adopters could not name what version they ran. Wave 1 dispatch needs a version anchor for support requests, rollback, and changelog reads.
-
-Per bassclef-upstream#614 (WU-1 of bet 2026-07-03b).
-
-## Scope
-
-Applies to:
-
-- `sunj-labs/bassclef-upstream` — the source-of-truth repo. Tags here name the cut point substrate work shipped from.
-- `sunj-labs/bassclef` — the public downstream repo adopters pull from. Tags here name what adopters see.
-
-Both repos carry the same tag name for each release. Adopters read the public tag; operators read the upstream tag when tracing "which cut point shipped this rule / hook / skill". Per ADR-039.
-
-## When tags cut
-
-Tags cut in two moments:
-
-1. **Upstream tag** — after the operator merges the `/release` PR on public bassclef. The `scripts/release-to-bassclef.sh` script tags upstream at the source SHA (currently `release-YYYY-MM-DD-<sha>` marker); this standard adds the semver tag on top.
-2. **Public tag** — immediately after the upstream tag lands, on the merged commit of the `/release` PR in public bassclef.
-
-Tags never cut mid-work. The `/release` dispatch is the single trigger.
-
-## Semver policy
-
-Bassclef follows semver (major.minor.patch). Each level names a kind of change adopters see.
-
-| Level | When it bumps | Examples |
-|---|---|---|
-| **major** | Adopter-observable breaking change | Repo rename, tier vocabulary change, an ADR-031-level break |
-| **minor** | New substrate surface adopters can call on | New rule, hook, skill, standard, luminary, agent |
-| **patch** | Cure or hardening; no new surface | Bug fix, wordlist amendment, refactor, doc rewrite |
-
-When a release carries multiple change classes, the highest applies. Two new rules plus a bug fix cut a minor bump.
-
-The first tag is `v0.1.0` — bassclef has shipped for 8 months plus 1500+ commits and holds ADR-035 substrate-as-system, which is stable enough to name.
-
-## Release notes
-
-Every release page carries release notes grouped by the substrate surfaces the release touched. The `/release` SKILL drafts these from the PRs the release picked up.
-
-Group order:
-
-1. **Rules** — new or amended
-2. **Hooks** — new or refactored
-3. **Skills** — new, amended, or retired
-4. **Standards** — new or amended
-5. **Luminaries** — new full or stub
-6. **Agents** — new or amended
-7. **ADRs** — new decisions
-8. **Cures** — bug fixes + hardening + wordlist amendments
-9. **Sync + release + install** — infrastructure changes that reach adopters at session-start
-
-Each group carries one line per PR. Links to the PR + the source ticket.
-
-Release notes render in three places:
-
-- **GitHub Release page** — the primary surface adopters see
-- **Public bassclef README banner** — a one-line "Latest: v0.1.0 shipped 2026-07-04" link at the top
-- **Adopter changelog** — `docs/changelog.md` in public bassclef; append-only
-
-## Where tags live
-
-- Bassclef-upstream: `v0.1.0`, `v0.1.1`, `v0.2.0`, ... cut on `main` at the release commit
-- Public bassclef: same tag names cut on `main` at the merged release commit
-
-Tags are signed only when the operator's GPG key is present. Unsigned tags still ship; signing is a follow-on when the release pipeline runs on the operator's own machine.
-
-Prior release-marker tags (`release-YYYY-MM-DD-<sha>`) stay in place. They serve a different purpose — naming which SHA a release PR built from. Semver tags name what adopters get. Both patterns co-exist.
-
-## Rollback
-
-When a release breaks adopters, the operator can:
-
-1. Roll adopter setups back — `bassclef-sync` in an adopter checkout can point at the prior tag via `.bassclef-source.json` `git_ref` field
-2. Issue a patch tag with the cure — `v0.1.1` fixes what `v0.1.0` broke
-3. Yank the tag only as a last resort — leaves a hole in the version sequence and confuses adopters running the yanked version
-
-Prefer forward-fix (option 2) over rollback (option 1) over yank (option 3).
-
-## Anti-patterns
-
-These shapes fail this standard.
-
-**Cut a tag between releases.** Tags cut only on `/release` dispatch. Every tag names an adopter-observable release moment.
-
-INSTEAD: hold cure work in `main` until the next release window. Batch multiple cures into one patch tag.
-
-**Bump minor for a cure.** Cures without a new surface stay in patch. Minor bumps signal "new stuff adopters can call on".
-
-INSTEAD: hold the standard bar. When in doubt, bump patch; adopters read patch bumps as "safe to pull".
-
-**Ship a release with no release notes.** Every tag carries a release page with the group-shape breakdown.
-
-INSTEAD: the `/release` SKILL drafts the notes from the PRs the release picked up. If drafting fails, the release blocks until the operator writes the notes by hand.
-
-**Different tags on upstream vs public.** The two repos carry the same tag names. Adopters read public; operators trace back through matching upstream tags.
-
-INSTEAD: `/release` cuts both tags in the same dispatch, with the same name.
-
-## Override
-
-`SKIP_RELEASE_TAG=1 <command>` for `/release` dispatches that ship substrate without a version bump (rare — mainly for infrastructure fixes that ship before the next planned release window). Logged via trace-helper.
-
-INSTEAD of overriding routine work: hold the change until the release window. The discipline compounds across every adopter version pin.
-
-## Composes with
-
-- `.claude/skills/release/SKILL.md` — the operator surface that dispatches releases + cuts tags
-- `scripts/release-to-bassclef.sh` — the release pipeline this standard adds tag cutting to
-- ADR-029 — release pipeline (parent threadhitecture)
-- ADR-039 — bassclef-upstream vs public bassclef tagging split (this standard's ADR)
-- ADR-031 — we-don't-break-adopters (major bumps signal what would otherwise break adopters)
-- `.claude/rules/we-dont-break-adopters.md` — sister discipline; ADR-031 methodology layer
-- `@luminary linus-torvalds` — kernel version discipline (stable API contract per tag)
-- `@luminary frederick-brooks` — conceptual integrity across versions
-- `@luminary michael-nygard` — stability patterns (the release page IS the adopter-facing steady state)
-
-## Refs
-
-- bassclef-upstream#614 — this standard's filing ticket
-- Parent goal: `docs/iteration-bets/2026-07-03b-wave-1-launch-prep.md` WU-1
-- Parent roadmap: `docs/roadmaps/2026-06-28-early-adopter-launch.md` Phase XI Batten Down
-- ADR-029 release pipeline (Sanjay P. + Claude Opus, 2026-06)
-- ADR-039 release tagging split (this bet WU-1)
-
-## Retirement condition
-
-This standard retires only if bassclef stops shipping to adopters. Semver policy may evolve (e.g., stricter major-bump criteria); the discipline of tagging every release survives.

package/substrate/standards/reserved-skill-names.md

@@ -1,120 +0,0 @@
----
-tier: lite
-description: Bassclef MUST NOT create a skill with a name that Claude Code (the harness) already claims.
----
-
-# Reserved Skill Names
-
-Bassclef MUST NOT create a skill with a name that Claude Code (the
-harness) already claims. Operator typing `/<name>` resolves to the
-harness built-in, not bassclef's skill — producing silent behavioral
-drift.
-
-This file is the **list**. The rule that references it lives at
-`.claude/rules/reserved-skill-names.md`. Separating the two lets the
-list evolve with Claude Code releases without churning the rule.
-
-## Reserved names (as of 2026-04-18)
-
-Source: Claude Code harness built-in skill registry observed in-session.
-
-### Claude Code operational built-ins
-
-These are always resolved by the Claude Code harness and will never
-reach bassclef's skill registry:
-
-| Name | Purpose (per harness) |
-|------|----------------------|
-| `update-config` | Configure Claude Code harness via settings.json |
-| `keybindings-help` | Keyboard shortcut customization |
-| `simplify` | Review changed code for reuse/quality |
-| `fewer-permission-prompts` | Scan transcripts + prioritize allowlist |
-| `loop` | Recurring prompt / slash-command execution |
-| `schedule` | Scheduled remote agents (triggers) |
-| `claude-api` | Build/debug Claude API / SDK apps |
-
-### Claude Code "FEY" remote built-ins
-
-Sourced from `FEY` array in Claude Code `cli.js`:
-
-| Name | Purpose (per harness) |
-|------|----------------------|
-| `autopilot` | Autopilot mode |
-| `bugfix` | Remote bugfix workflow |
-| `dashboard` | Remote dashboard |
-| `docs` | Docs workflow |
-| `investigate` | Investigation workflow |
-
-### Claude Code init/setup
-
-| Name | Purpose (per harness) |
-|------|----------------------|
-| `init` | Initialize a new CLAUDE.md file |
-| `review` | Review a pull request |
-| `security-review` | Security review of pending changes |
-
-## What the rule enforces
-
-Any new `.claude/skills/<name>/SKILL.md` introduced by a PR MUST NOT
-use a name from this list. The pre-commit hook blocks commits that
-add a colliding directory. See `.claude/rules/reserved-skill-names.md`
-for full enforcement semantics.
-
-## Updating this list
-
-Triggers that justify adding a name:
-
-- New Claude Code release adds a slash-command built-in not listed here
-- Observed behavioral drift where a bassclef skill name was silently
-  overridden by the harness
-- Upstream Anthropic documentation announces a reserved prefix or name
-
-Process:
-
-1. Verify the collision exists (type `/<name>` in a clean session,
-   observe the harness responds without reaching bassclef)
-2. Add the row to the table above with purpose description
-3. Bump the date in the heading ("as of YYYY-MM-DD")
-4. File a bassclef issue if any existing bassclef skill uses the
-   newly-colliding name — that skill MUST be renamed before the rule
-   addition merges
-
-## Verification
-
-Test case for the pre-commit hook lives at
-`.claude/hooks/tests/reserved-names-test.sh`. Run via:
-
-```bash
-bash .claude/hooks/tests/reserved-names-test.sh
-```
-
-Tests:
-- Hook blocks commit staging `.claude/skills/docs/SKILL.md` (collision)
-- Hook permits commit staging `.claude/skills/docs-v2/SKILL.md` (no collision)
-- Hook respects `SKIP_RESERVED_NAMES=1` override
-
-## Override
-
-Rare case: bassclef deliberately wants to shadow a harness built-in
-(e.g., wraps or replaces the built-in). Requires:
-
-1. Explicit operator confirmation in PR body
-2. `SKIP_RESERVED_NAMES=1` env var on the commit
-3. Rationale documented in the skill's SKILL.md frontmatter
-
-The override exists because rules should not be unjudgeable — but the
-friction is deliberately high.
-
-## Relationship to other standards
-
-- `standards/substrate-config-schema.md` — external resource reference
-  conventions (orthogonal)
-- `.claude/rules/reserved-skill-names.md` — the rule that reads this list
-- `.claude/hooks/pre-commit-gate.sh` — mechanical enforcement
-
-## When to retire this list
-
-If Claude Code ships a mechanical API to query reserved names at
-runtime (e.g., `claude --list-builtins --json`), this static list can
-be replaced with a dynamic check. Until then, the list is the
-interface.